by Prescription Only
Now Available by Prescription Only
A New Path Forward.
Canvas Dx is a Software as a Medical Device (SaMD) that aids physicians in diagnosing autism spectrum disorder (ASD) in young children1,2,3
Canvas Dx harnesses clinically validated artificial intelligence (AI) technology to aid physicians in diagnosing ASD in children between the ages of 18 and 72 months who are at risk of developmental delay4,5
Canvas Dx has the potential to facilitate early diagnosis in children with ASD4,5
How does Canvas Dx work?
Canvas Dx incorporates 3 separate, user-friendly inputs:
A parent/caregiver questionnaire that asks about the child’s behavior and development collected via a parent/caregiver facing app3
A questionnaire completed by a video analyst who reviews two videos of the child recorded by the parent/caregiver3
A HCP questionnaire completed by a physician* who meets with the child and a parent/caregiver, collected via a health care provider portal3
*One of Canvas Dx’s device inputs is a 13-15 item age-dependent health care provider (HCP) questionnaire collected via a health care provider portal. Cognoa has contracted with a pediatric care provider to offer the option to have a qualified HCP complete the HCP questionnaire via a video visit with the caregiver and child, with the goal of allowing for a streamlined experience.
The Canvas Dx algorithm then evaluates all 3 inputs, generating a device output that the PCP utilizes in combination with their clinical judgement3,6
How AI Works
AI uses pattern recognition to uncover clinically meaningful relationships that exist between signs, symptoms, and behaviors. In addition, AI approaches are suitable for aiding medical diagnosis. They can unlock clinically relevant information hidden in healthcare data and identify complex non-linear patterns between patient features and diagnosis.33-47
How Canvas Dx Works
Canvas Dx harnesses the power of artificial intelligence (AI) to aid primary care physicians (PCPs) in the identification of autism spectrum disorder (ASD).29-32 As an AI / machine learning algorithm, Canvas Dx can be optimized to improve accuracy by training on new data.
The Canvas Dx algorithm (v2) was optimized by training on new data
and has the following performance metrics:
- Negative Predictive Value (NPV) of 95.6%
- Positive Predictive Value (PPV) of 87.5%
- Determinate rate (rate for Canvas Dx to provide either a positive or negative for ASD result) of 66.5%
Good machine learning practices were followed in the algorithm update, as the industry standard in the field of AI/ML.
When inputs are insufficient for rendering a high-confidence result, patients receive an "indeterminate" output. Canvas Dx algorithm (v2) has an indeterminate rate of 33.5%; data analysis of the children in this indeterminate group found that 95.9% were identified as having one or more neurodevelopmental conditions
Pivotal Study Results
The Canvas Dx pivotal study measured the performance of the Device with algorithm v1 in aiding the diagnosis of autism spectrum disorder (ASD), compared to specialist evaluation using DSM-5 criteria.48,49
The accuracy of Canvas Dx was assessed in comparison to the standard diagnostic approach (specialist evaluation and diagnosis) in a multisite, prospective, double-blinded, active comparator cohort study. The study included 425 children, aged 18-72 months, with parental or PCP concern for developmental delay.48,49
How Canvas Dx Helps
When used in conjunction with a physician's clinical assessment, Canvas Dx can accurately aid diagnosis starting at 18 months of age during a critical window in a child’s neurodevelopment. 10-12, 14, 23, 24
By helping PCPs diagnose or rule out ASD, remotely or in person, Canvas Dx has the potential to expand the group of diagnosing physicians and may allow for more efficient specialty referrals.25,26,27
Use of Canvas Dx may allow children with ASD to be diagnosed over 1.5 years earlier than the current average age of diagnosis49,54,55
The average age of children with a positive output in the Canvas Dx pivotal trial was 2.8 years, whereas the current average age of diagnosis is 4 years, 2 months.54,55